Skip to main content
Key Specialties
Cardiology Diabetology Endocrinology Gastroenterology Geriatrics and Gerontology Gynecology and Obstetrics Hematology Infectious Disease Internal Medicine Nephrology Neurology Oncology Pediatrics Respiratory Medicine Rheumatology Surgery
Congress Coverage
2024 ACC Congress 2023 ESMO Congress 2023 EASD Congress 2023 ERS Congress 2023 ESC Congress 2023 EHA Congress 2023 ASCO Annual Meeting
Clinical Collections
Advances in Alzheimer’s

Keynote webinar | Spotlight on sleep in brain health

LIVE: Thursday 2nd May 2024, 18:00-19:30 (CEST)

Quality sleep is essential for health. But what happens to our brains when sleep patterns are disturbed? Join our experts to explore the interplay between sleep disruption and neurological diseases, and the questions that you need to be asking your patients to help you prevent the harmful effects of sleep deprivation.
ADVANCED SEARCH
Log in
Top
Health Services and Outcomes Research Methodology
Published in: Health Services and Outcomes Research Methodology 1-2/2011

01-07-2011

Systematic self-report bias in health data: impact on estimating cross-sectional and treatment effects

Author: Sebastian Bauhoff

Published in: Health Services and Outcomes Research Methodology | Issue 1-2/2011

Login to get access

Abstract

This paper examines the effect of systematic self-report bias, the non-random deviation between the self-reported and true values of the same measure. This bias may be constant or variable, and can mislead empirical analyses based on descriptive statistics, program evaluation and instrumental variables estimation. I illustrate these issues with data on self-reported and measured overweight/obesity status, and BMI, height and weight z-scores of public school students in California from 2004 to 2006. I find that the prevalence of overweight/obesity is 2.4–7.6% points lower in self-reported data relative to measured data in the cross-section. A school nutrition policy changed the bias differentially in the treatment and control groups so that program evaluations could find spurious positive or null impacts of the intervention. Potential channels for this effect include improved information and stigma.
Footnotes
1
The prevalence of overweight is measured as the percentage of individuals above a specific cutoff in the distribution of body-mass index (BMI). BMI is calculated as (weight in kg)/(height in m)\(^2\). For children the cutoffs vary by age and gender (Vidmar et al. 2004).
 
2
The CHKS is based on the California Student Survey along with items from the Youth Risk Behavioral Survey (YRBS). It is required of districts that accept funds through the federal Title IV Safe and Drug-Free Schools and Communities (SDFSC) or the state Tobacco-Use Prevention Education (TUPE) programs.
 
3
Since 2001 the PFT is administered annually between February 1 and May 30 but the actual survey dates are not available in the data. I use April 1 as midpoint in determining the age cutoffs. I also only consider students in “pure” middle (offering at most grades 6–8) and high schools (offering grades 9 and higher). Most Californian schools meet these criteria. Less than 0.5% of the data are biologically implausible by the CDC standards, and the results reported below are similar for the full sample.
 
4
The cut-off is recommended by the Childhood Obesity Working Group of the International Obesity Taskforce (Vidmar et al. 2004). For some students weight is reported twice in the PFT, in the aerobic capacity and the body composition test. Where the reported weights disagree I use the weight from the body composition section. From 2002/3 onward the PFT also reports BMI which I use to infer missing values in weight with help of the height measure. When both weight measures are missing I use this calculated weight. If the composite BMI and the body composition weight are both missing I use the weight from the aerobic capacity test as last resort.
 
5
Values for California exclude five other districts that also implemented nutrition guidelines (Samuels et al. 2006).
 
6
Hausman specification tests reject the random intercept models but results are comparable with the fixed-effects results shown here. Since clustering on the district level increases the standard errors on the coefficients of interest due to negative intracluster correlation, I report the more conservative unclustered errors for all models.
 
7
Since I cannot link the self-reported and measured data for individual students, I am unable to see the effects on height and weight for students near the overweight/obesity cutoff.
 
8
This information effect may be particularly important for adolescents since their physique is changing rapidly and truthful reporting of their overweight status would require constant updating.
 
Literature
Bauhoff, S.: The Effect of School Nutrition Policies in California on Dietary Intake and Obesity: A Synthetic Control Approach. Working Paper (2010)
Bound, J., Brown, C., Mathiowetz, N.: Measurement error in survey data. In: Heckman, J., Leamer, E. (eds.) Handbook of Econometrics, vol. 5, chap. 59, pp. 3705–3843. Elsevier, Amsterdam (2001)
Brener, N.D., Eaton, D.K., Lowry, R., McManus, T.: The association between weight perception and BMI among high school students. Obes. Res. 12, 1866–1874 (2004)PubMedCrossRef
Brener, N.D., Mcmanus, T., Galuska, D.A., Lowry, R., Wechsler, H.: Reliability and validity of self-reported height and weight among high school students. J. Adolesc. Health 32, 281–287 (2003)PubMedCrossRef
Cawley, J.: Rational Addiction, the Consumption of Calories, and Body Weight, Ph.D. thesis (1999)
Cawley, J., Meyerhoefer, C., Newhouse, D.L.: The impact of state physical education requirements on youth physical activity and overweight. Health Econ. 16, 1287–1301 (2007)PubMedCrossRef
CDC: YRBS 2007 Data User Manual. Technical Report, Centers for Disease Control and Prevention (2007)
Ezzati, M., Martin, H., Skjold, S., Hoorn, S.V., Murray, C.J.L.: Trends in national and state-level obesity in the USA after correction for self-report bias: analysis of health surveys. J. R. Soc. Med. 99, 250–257 (2006)PubMedCrossRef
Puhl, R.M., Latner, J.D.: Stigma, obesity, and the health of the nation’s children. Psychol. Bull. 133, 557–580 (2007)PubMedCrossRef
Puhl, R.M., Latner, J.D.: Weight bias: new science on an significant social problem. Obesity 16, S1–S2 (2008)PubMedCrossRef
Samuels, S.E., Craypo, L., Boyle, M., Stone-Francisco, S., Schwarte, L.: Improving School Food Environments Through District-Level Policies: Findings from Six California Case Studies. Technical Report, Samuels and Associates (2006)
StataCorp: Stata Statistical Software: Release 11 (2009)
Vidmar, S., Carlin, J., Hesketh, K., Cole, T.: Standardizing anthropometric measures in children and adolescents with new functions for egen. Stata J. 4, 50–55 (2004)
Metadata
Title
Systematic self-report bias in health data: impact on estimating cross-sectional and treatment effects
Author
Sebastian Bauhoff
Publication date
01-07-2011
Publisher
Springer US
Published in
Health Services and Outcomes Research Methodology / Issue 1-2/2011
Print ISSN: 1387-3741
Electronic ISSN: 1572-9400
DOI
https://doi.org/10.1007/s10742-011-0069-3

Other articles of this Issue 1-2/2011

Health Services and Outcomes Research Methodology 1-2/2011 Go to the issue

OriginalPaper

On stratified sampling and ratio estimation in medicare and medicaid benefit integrity investigations

OriginalPaper

Estimating treatment effects on healthcare costs under exogeneity: is there a ‘magic bullet’?

OriginalPaper

Well-being and obesity of rheumatoid arthritis patients

OriginalPaper

Genes as instruments for studying risk behavior effects: an application to maternal smoking and orofacial clefts